A hedge fund may been defined as a pool of private capital used to leverage an investment portfolio structured as a limited partnership. A General Partner (Investment Fund Manager) is expected to have sophisticated portfolio management practices; limited partners (investors) are generally high net-worth individuals and institutions that act as silent partners. Hedge funds often have restricted liquidity, since they cannot be traded in the open market. Moreover, hedge funds are not required to report their assets or returns to the Securities & Exchange Commission (SEC) or to any other entity whose information is made available to the public. And hedge fund advertising is restricted. Thus, unless a hedge fund voluntarily reports its existence and performance to some database, the public may be unable to learn of either.
In recent years the term “hedge fund” has been used broadly to refer to any type of alternative investment strategy (“AIS”). UBS Warburg Global Equity Research defines AIS as “an asset class that seeks to generate absolute positive returns by exploiting market inefficiencies while minimizing exposure and correlation to traditional stock and bond investments.” See Alexander M. Ineichen CFA, In Search of Alpha—Investing in Hedge Funds, UBS Warburg Global Equity Research, October 2000, page 6 (hereinafter cited as “Alpha”). The term “hedge fund” is used herein, as in the art, interchangeably with the terms alternative investment strategy and AIS.
Investable Index Funds in Traditional Asset Classes
An index fund (also known as a passively-managed fund) seeks to match the investment performance of a specific stock or bond benchmark index. This definition of an index fund can be found, for example, at http://www.vanguard.com, hereinafter referred to as “Vanguard.” The Vanguard Group is one of the world's largest index fund managers for traditional stock and bond markets. Instead of actively trading securities in an effort to beat the market, an index fund manager simply holds all, or a representative sample, of the securities in the index. In contrast, an active fund manager buys and sells securities regularly in pursuit of maximum gain.
Investors typically place money in an index fund to replicate the performance of a market. Mature, liquid marketplaces have a host of well-defined benchmarks or indices, which corresponding index funds seek to replicate. For example, the S&P 500, Russell 2000, and Wilshire 5000 all are benchmarks whose performances are mimicked by index funds. The benchmark represents the passive alternative—a portfolio that faithfully replicates asset class performance.
However, “beating the market, as described by a fair benchmark, constitutes the foremost object of an investment manager.” See David F. Swensen, Pioneering Portfolio Management, The Free Press, 2000, page 300, cited hereinafter as “Swensen.” And although everyone would like to outperform the market, Vanguard believes “it's more difficult than you might think . . . [because] investing is a Zero-Sum Game, mutual fund costs diminish returns, and financial markets are efficient.”See Vanguard.
David Swensen, Chief Investment Officer of the Yale University endowment, echoes the problems of engaging in an active strategy:                In spite of the daunting obstacles to active management success, the overwhelming majority of market participants choose to play the loser's game. Like the residents of Lake Wobegon, who all believe their children to be above average, all investors believe their active strategies will produce superior results. The harsh reality of the negative sum game dictates that in aggregate, active managers lose to the market by the amount it costs to play, in the form of management fees, trading commissions, and dealer spread. Wall Street's share of the pie defines the amount of performance drag experienced by the would-be market beaters.Swensen, page 6.        
Because of the difficulty in adding value through active management, many investors in equity and debt markets have chosen to employ passive strategies. Popularized by Vanguard and Barclays Global Investors, index funds use a statistical sampling of securities in a defined market to replicate the performance of that market.
Lack of Index Funds in Alternative Investment Strategies
Despite having years of experience purchasing stocks and bonds, institutional and high net worth investors have only recently added alternative investment strategies to their investment portfolios. In contrast to traditional stock and bond portfolios, alternative investment strategies do not have index funds that offer investors a passive alternative to actively managed portfolios. Not only is the market or benchmark difficult to define, creating an index fund to encompass a defined market has an assortment of its own difficulties. As a result, surveyors of the alternative investment arena have assumed that investable indices in alternative strategies cannot exist.
For example, Swensen believes that “investors in alternative asset classes must pursue active management since market returns do not exist in the sense of an investable passive option.” Swensen even defines the term “absolute return” to impress that lack of an index: “Absolute return investments seek to generate high levels of returns, independent of market results, contrasting with the relative, benchmark-beating gains pursued by active marketable security managers.” See Swensen, pages 204–05.
As another example, UBS Warburg makes the assumption that alternative investment strategies are de facto measured based on absolute performance. In comparing hedge funds to mutual funds, UBS Warburg claims “hedge funds are measured based on absolute performance, mutual funds are usually measured based on relative performance.” See Alpha, page 54.
Existing Performance Benchmarks
A prerequisite for developing an index fund is the definition of the applicable performance benchmark. Currently, investors in hedge funds measure performance against an absolute return hurdle or against other alternatives (hedge fund peers). However, as explained below, neither benchmark approach can be replicated in an index fund.
The early adopters of alternative investment strategies measured active managers against an absolute hurdle—for example, an annual 8% real return. While investors would relish the certainty of investing in a known 8% real return, no vehicles exist to do so. Thus, an absolute return benchmark as high as the aspirations for this asset class cannot be turned into an index fund for passive investors.
More recently, consulting organizations such as the Hennessee Group LLC, Hedge Fund Research Inc., and Tremont Advisers, Inc. have constructed peer indices aggregating hedge fund performance from databases. While such peer indices provide a numerical benchmark, the indices suffer from measurement biases that are rampant in databases of historical hedge fund returns. As noted by William Fung and David A. Hsieh (in “Benchmarks of Hedge Fund Performance: Information Content and Biases,” Research paper, Fuqua School of Business, Duke University, February 2001, cited hereinafter as “Hedge Fund Benchmarks”, pages 2–4), hedge fund databases suffer from survivorship bias and selection bias.
Survivorship bias arises when a sample of hedge funds includes only funds that are operating at the end of the sampling period and thus excludes funds that have ceased operations during the period. From the universe of all hedge funds, those that ceased to exist before database vendors started data collection, ceased operations (“dead funds”), or simply exited a database, whether voluntarily or not (“defunct funds”), all cause tracking error between the peer index and the universe of hedge funds. Academics have estimated that survivorship bias in hedge fund peer indices averages 2% per annum. See Gaurav S. Amin and Harry M. Kat, “Welcome to the Dark Side: Hedge Fund Attrition and Survivorship Bias Over the Period 1994–2001,” working paper, Dec. 11, 2001, page 2.
Selection bias further degrades the accuracy of peer indices. Hedge funds that enter databases typically do so to attract outside investor capital. Presumably, only those funds that have “good” performance and are looking to attract new investors want to be included in a database. Therefore, hedge funds in a database tend to have better performance records than those that are excluded. In contrast, successful “closed” hedge funds with no reason to attract outside capital may have no need to be included in databases. This effect would cause databases to understate industry performance. Both self-selection biases create errors in the historical reporting of peer indices. See, e.g., Hedge Fund Benchmarks, pages 7–9. As a result of such measurement biases, peer indices do not accurately reflect the hedge fund market.
Due to the inherent biases in existing peer indices, Fung and Hsieh come to the conclusion that “if we want to estimate the investment experience of hedge funds, why not look directly at the experience of the hedge fund investors themselves.” They recommend looking at the investment records of funds-of-hedge funds (FOFs):                Survivorship bias in FOFs' returns relative to the universe of all hedge fund investors is less severe than using individual hedge funds. This is because FOFs, through the natural process of diversification, have inadvertently minimized the measurement errors that may arise. Also, the question of selection bias is much more muted in the case of FOFs due to the fact that all individual hedge funds are likely to have some FOF investors irrespective of whether they report performance to any hedge fund database . . . we would argue that using FOFs as building blocks for hedge fund performance benchmarks is a better alternative to individual funds.Hedge Fund Benchmarks, pages 10–12.        
Although Fung and Hsieh reach the conclusion that funds of funds should be used as “building blocks” for hedge fund performance benchmarks, they do not disclose creating an investable index fund based on an index of funds of funds. Consequently, they also fail to disclose how such a fund might be created and managed. In fact, they implicitly assume and suggest that the fees for such a fund would make it impractical (see page 12 of Hedge Fund Benchmarks, where returns for broad-based indices (that is, indices of hedge funds) were estimated to be “roughly the same” as for FOF indices). This implies that the extra layer of fees that an index fund based on an index of funds of funds would entail would cause such an index fund to be noncompetitive.
Hedge Fund Benchmarks discloses that certain commercial databases have posted indices based on funds of funds. However, they have not managed to create a fund based on those indices, nor does Hedge Fund Benchmarks disclose a practical method of creating and managing such a fund. Implementation of a portfolio for such a fund is far from trivial, as shown below. For example, a clearly defined set of rules for the inclusion of certain funds of funds in the portfolio (e.g., size, track record, positions, concentration) is required. The indices mentioned in Hedge Fund Benchmarks simply attempt to include all of the FOFs in their database, resulting in tracking errors similar to those in hedge fund databases. Further, there is no suggestion in Hedge Fund Benchmarks that a fund based on those indices would not cost an investor in the index fund an extra layer of fees beyond the hedge fund manager fees and funds of funds manager fees.
Index Fund Attempts
A few financial services firms (notably Zurich Capital Markets and Morgan Stanley) have attempted to create an investable hedge fund index through investment in a portfolio of hedge funds. In its marketing materials, Zurich describes its methodology as follows: “The Indices' methodology begins by identifying a set of candidate hedge fund managers who are potential members of any of the Indices . . . the candidates include all managers who actively report data to public databases, as well as other manager who are non-reporters but who are identified through other means.”
Creating an index fund through investment in other hedge funds has a series of theoretical and practical problems. First, the universe of all hedge funds is not known. Although commercial databases collect information on funds, some of the best and worst hedge funds do not appear in the databases (for reasons described above). Without knowledge of the characteristics of the universe of hedge funds, creating a competent statistical sample is problematic.
Second, many leading hedge funds are closed to new investors. Longstanding survivors in the business that do not accept new investment capital are outside of the universe of opportunities available to a new fund of funds. Many of these funds drive the performance that institutional investors desire when allocating capital to alternative investment strategies.
Third, investors in an index fund want immediate diversified exposure to the asset class. A new hedge fund index may have difficulty building an appropriate diversified portfolio.
Fourth, the administration of investment in hundreds of funds (the number likely required for a statistically representative sample) would be a logistical nightmare. The universe of hedge funds is very large (estimated at over 6,000 funds) and complex (one industry-leading database currently lists 25 separate strategies for hedge fund investments).
Creating a rigid process to maintain an index fund is extraordinarily difficult with an ever-changing industry. Given these constraints, perhaps it is not surprising that the vehicles introduced by Zurich and Morgan Stanley have thus far failed to be embraced by outside investors.
There is thus a need for a hedge fund index based on a portfolio of funds of funds that solves the problems described above. Such an index fund would preferably: (a) mimic the experience of hedge fund investors; (b) provide immediate exposure to the entire market; (c) comprise an important component of the market; (d) be systematically replicable; and (e) provide a low-cost, passive alternative to fund-of-funds investments.
Mimicking the Experience of Hedge Fund Investors
In order to diversify across strategies, investors in hedge funds create a portfolio of multiple managers. An index fund constructed by investment in a portfolio of FOFs preferably provides an investable passive alternative capable of replicating the experience of hedge fund investors broadly.
Immediate Exposure to the Entire Market
An index fund formed by direct hedge fund investment cannot offer exposure to currently closed hedge funds. In contrast, a fund formed by investment in funds of funds is more likely to gain access to closed funds, as existing funds of funds are often clients of closed funds.
Important Component of the Market
According to UBS Warburg, funds of funds manage 20–25% of all hedge fund assets. See Alexander M. Ineichen CFA, The Search for Alpha Continues—Do Fund of Hedge Funds Managers Add Value?, UBS Warburg Global Equity Research, September 2001, page 28. While industry researchers estimate a universe of 6,000 hedge funds, the Altvest database counts only 375 funds of funds, and only 72 with over $100 million in assets.
Systematic Replication Achievable
As opposed to administering a portfolio with hundreds if not thousands of hedge funds, investment in a far smaller number of funds of funds is readily achievable and can offer the same exposure as a portfolio of direct investments in hedge funds. Fung and Hsieh note that “it takes fewer FOFs to deliver the same diversification benefit as individual hedge funds, and FOFs are likely to have a more uniform redemption policy than individual funds.” Hedge Fund Benchmarks, page 10.
Low Cost Passive Alternative
One perceived problem with the creation of a portfolio of FOFs is the addition of yet another layer of fees on top of those of the hedge fund manager and the FOF manager. It would be preferable to maintain fee to investors at the same fee level as would occur if the investors placed capital in a FOF.